Take-home Exercise 4

In this Take-home Exercise, I will explore to reveal the daily routine of two selected participants of the city of Engagement, Ohio USA.

Task

Challenge 2: Patterns of Life considers the patterns of daily life throughout the city. You will describe the daily routines for some representative people, characterize the travel patterns to identify potential bottlenecks or hazards, and examine how these patterns change over time and seasons.

In Challenge 2, you will use visual analytic techniques to address these questions:

• Participants have given permission to have their daily routines captured. Choose two different participants with different routines and describe their daily patterns, with supporting evidence. Limit your response to 10 images and 500 words.

Overview

In this take-home exercise, ViSiElse and other appropriate visual analytics methods are used to reveal the life patterns of the city of Engagement, Ohio USA while addressing the questions stated in the Task section.

The data are processed by using appropriate tidyverse family of packages.

Sketch of Proposed Design

The picture below shows a sketch of the initial design proposed.

Installing & Launching R Packages

Before we get started, it is important for us to ensure that the required R packages have been installed. If yes, we will load the R packages. If they have yet to be installed, we will install the R packages and load them onto R environment.

The chunk code below will do the trick.

packages = c('tidyverse', 'data.table', 'knitr', 'ViSiElse', 'kableExtra', 'lubridate', 'ggplot2', 'ggiraph', 'plotly', 'DT', 'patchwork','crosstalk')

for(p in packages){
  if(!require(p, character.only = T)){
    install.packages(p)
  }
  library(p, character.only = T)
}

Importing Data

The code chunk below imports ActivityLogs files from the rawdata folder into R by using list.files(), read_csv() of readr package, and map_df() of purrr package, and and save it as a tibble data frame called logs.

memory.limit(size=160000)

logs <- list.files(path = "./rawdata/ActivityLogs/",
                   pattern = "*.csv", 
                   full.names = T) %>% 
  map_df(~read_csv(., 
                   col_types = cols(.default = "c"))) 

Data Wrangling

During our data exploration, we realized some of the participants starts off with unstable financial status while some having stable financial status. As such, we will choose 1 participant each, with partipantId 4 from the unstable financial status group and partipantId 7 from the stable financial status group, and to look after their differing routines.

The code chunk below limits our logs data to the only selected participants using filter() function.

logs_two_p <- logs %>%
  filter(participantId == '4' | participantId == '7')

Data frame logs_two_p is saved in RDS format to avoid uploading large files to Git.

saveRDS(logs_two_p, 'data/logs_two_p.rds')

logs_two_p <- readRDS('data/logs_two_p.rds')

We will use kable() to review the structure of the data frame and kable_classic of kableExtra has been applied to ensure the full width of the table can be preserved.

kable(head(logs_two_p), format = "html", table.attr = "style='width:40%;'" ) %>% 
  kable_classic(full_width = T, position = "center", )

timestamp	currentLocation	participantId	currentMode	hungerStatus	sleepStatus	apartmentId	availableBalance	jobId	financialStatus	dailyFoodBudget	weeklyExtraBudget
2022-03-01T00:00:00Z	POINT (976.2409614204214 4574.575079082071)	4	AtHome	JustAte	Sleeping	194	-681.6505879983722	177	Unstable	20	0
2022-03-01T00:00:00Z	POINT (-1023.8165705255449 1578.3713681439597)	7	AtHome	JustAte	Sleeping	97	683.8632217708105	270	Stable	20	1164.0690586556877
2022-03-01T00:05:00Z	POINT (976.2409614204214 4574.575079082071)	4	AtHome	JustAte	Sleeping	194	-681.6505879983722	177	Unstable	20	0
2022-03-01T00:05:00Z	POINT (-1023.8165705255449 1578.3713681439597)	7	AtHome	JustAte	Sleeping	97	683.8632217708105	270	Stable	20	1164.0690586556877
2022-03-01T00:10:00Z	POINT (976.2409614204214 4574.575079082071)	4	AtHome	JustAte	Sleeping	194	-681.6505879983722	177	Unstable	20	0
2022-03-01T00:10:00Z	POINT (-1023.8165705255449 1578.3713681439597)	7	AtHome	JustAte	Sleeping	97	683.8632217708105	270	Stable	20	1164.0690586556877

We will add date and year_month columns for further analysis.

logs_two_p$date <- date(logs_two_p$`timestamp`)
logs_two_p$year_month <- format(as.Date(logs_two_p$`timestamp`), "%Y-%m")

Let’s look at the different activities from currentMode:

[1] "AtHome"       "Transport"    "AtWork"       "AtRestaurant"
[5] "AtRecreation"

We will now calculate the duration of each activity in seconds and transform data using pivot_wider() of tidyr to suit the usage for visielse() visuals.

logs_daily_ss <- logs_two_p %>%
  group_by(date, participantId, currentMode) %>%
  summarise(count = n()) %>%
  mutate(ss = count * 5 *60) %>%
  ungroup() %>%
  select(date, participantId, currentMode, ss) %>%
  pivot_wider(names_from=currentMode, values_from=ss,
              values_fill=0)

date	participantId	AtHome	AtRestaurant	AtWork	Transport	AtRecreation
2022-03-01	4	54600	1200	28800	1800	0
2022-03-01	7	28200	2400	29100	21300	5400
2022-03-02	4	53700	1200	29100	2400	0
2022-03-02	7	27600	1200	29100	17100	11400
2022-03-03	4	33000	1200	29700	12300	10200
2022-03-03	7	27600	2400	29100	22500	4800

ViSiElse Plot

Before plotting, we will set an order to the activities.

logs_daily_ss <- logs_daily_ss[,c(1,2,5,4,3,6)]

visielse(logs_daily_ss, group = logs_daily_ss$participantId, informer = NULL, method = "cut", pixel = 1000)

-parameters 
 method           :  cut 
 grwithin         :   
 quantity         :  N 
 informer         :   
 tests            :  FALSE 
 threshold.test   :  0.01 
 pixel            :  1000 
 t_0              :  0 
-MATp             : 10 x 89 sparse Matrix of class "dgCMatrix" 
-L                : 0 x 0  data.frame 
-idsort           : 0 x 0 matrix 
-MATpsup          : 0 x 0 sparse Matrix of class "dgCMatrix" 
-idsup            : length  0 vector 
-Lsup             : 0 x 0  data.frame 
-colvect          : 2 x 1 matrix 
-BZL              : 0 x 0 sparse Matrix of class "dgCMatrix" 
-book             : 5 x 6  ViSibook 
-group            : length  900 factor 
-vect_tps         : length  89 vector 
-testsP           : length  0 vector 
-informers        : 0 x 0 matrix 

We realize that in general Participant 4 is spending less time in restaurant and at recreation than Participant 7, possibly due to his unstable financials status. This is more obvious in the difference in time spent for recreation where Participant 4 might want to restrict his non-necessity related spending in order to improve his financial outlook. This also aligns with the time spent at home where Participant * spent much more time at home than Participant 7. We also spot that both participants are having the same amount of time spent at work. As for transport perspective, Participant 4 is spending less time than Participant 7, potentially due to a closer location from home.

Static Trellis Plot

We have transformed the data accordingly and displayed a trellis plot partitioned by currentMode

qplot(year_month, hour, data = logs_two_p_count, color = participantId) +
  facet_wrap(~ currentMode) +
  labs(y= 'Hour', x= 'Time',
     title = "Trend for Respective Activities Over Time") +
  guides(color = guide_legend(title = "Participant Id")) +
  theme(axis.title.y= element_text(angle=0), axis.ticks.x= element_blank(), axis.text.x=element_blank(),
      axis.line= element_line(color= 'grey'))

We can see that despite a few of the irregular patterns, generally the time spent on each activity follow similar trends for both participant 4 and participant 7.

Interactive Trellis Plot

In order to add interactivity to the previous trellis plot, we will use ggplotly() of plotly and bscols() of crosstalk.

dd <- highlight_key(logs_two_p_count)

graph1 <- qplot(year_month, hour, data = logs_two_p_count, color = participantId) +
  facet_wrap(~ currentMode) +
  labs(y= 'Hour', x= 'Time',
     title = "Trend for Respective Activities Over Time") +
  guides(color = guide_legend(title = "Participant Id")) +
  theme(axis.title.y= element_text(angle=0), axis.ticks.x= element_blank(), axis.text.x=element_blank(),
      axis.line= element_line(color= 'grey'))

gg <- highlight(ggplotly(graph1),
                "plotly_selected")

crosstalk::bscols(gg, 
                  widths = c(12, 12),
                  DT::datatable(dd,
                                rownames = FALSE,
                                colnames = c('Year Month' = 'year_month',
                                             'Participant Id' = 'participantId',
                                             'Current Mode' = 'currentMode',
                                             'Hour' = 'hour'),
                                filter = 'top',
                                class = 'display'))